a16z on Protecting Little Tech: The Techno-Optimist AI Policy Agenda with Matt Perault, Head of AI Policy

Hello, and welcome back to The Cognitive Revolution. Today, I'm pleased to have Matt Perault, head of AI policy at Andreessen Horowitz, here to discuss a16z's approach to AI regulation, including their advocacy for AI policies that support experimentation and innovation by little tech startups, and their efforts to avoid common pitfalls like regulatory capture that could further entrench big tech incumbents. We cover a wide range of important topics, Starting with A16Z's core principle that regulating harmful AI use is a better approach than regulating the research and development process itself. We also cover the RAISE Act, sponsored by New York State Assembly member and recent guest Alex Boris, and the challenges of setting appropriate thresholds for regulation. Transparency requirements and what information companies should and shouldn't be required to disclose. And the pros and cons of California's SB 813, which could create a novel public-private regulatory framework that offers liability shields to companies in exchange for opt-in compliance. Overall, I was again struck by how much common ground exists between the a16z worldview and that of many in the AI safety community, myself included. Our techno-optimist credentials are perhaps best proven by the fact that we began to take both the upside and risks of AI seriously years before the technology itself really started to work. And many of us share a16z's generally libertarian politics, veneration of entrepreneurship, and fear of concentration of power. With that shared foundation in mind, I think it is clearly good that someone is working to ensure that AI capabilities don't become unduly concentrated in just a handful of companies. And Matt does raise some important considerations in this conversation, including the possibility that SB 813 might create an unfair advantage for incumbents by setting the compliance costs for liability protection so high that only the biggest companies are able to participate. I've not yet studied this issue in depth, but it does strike me as a worthwhile red teaming of the bill, and I definitely plan to raise that concern with some of the people who are advocating for SB 813 in an upcoming episode. Meanwhile, where important disagreements persist, and you will hear a few, it seems to me that they largely stem from disagreement on empirical questions. Such as whether scaling laws do or don't imply that only a small number of hyperscalers will push the frontiers of AI forward over the next couple of years, and more importantly, differing expectations about just how powerful AI could become and how soon that might happen. With those differences in mind, I am definitely more concerned than Matt is about the risks of secretive R&D processes, particularly if companies are allowed to double down on AI-powered machine learning research, to execute large-scale YOLO training runs without careful testing throughout the training process, and to continue deploying models internally without the same safety standards that have become industry best practice for public deployments. All this to me does seem to create the potential for surprising and potentially even irreversible harms for which prevention is really the only viable option. That said, I still came away from this conversation feeling very optimistic. That the overwhelming majority of non-ideological people who simply want what's best for everyone can converge on a shared understanding and smart policy ideas as more evidence comes in. I personally would love to see a research breakthrough that solves core AI safety and control issues on a technological level, and I would happily update my policy positions if that happens. On the other hand, it seems that mounting evidence of autonomous bad behaviour from AI systems also has the potential to change at least some accelerationist minds. Timelines, of course, might be quite short, so it's critical to keep tracking all these developments in real time. And I look forward to continuing to bring people of very different perspectives together to grapple with the latest evidence in good faith as we move through this critical period. As always, if you're finding value in the show, we'd appreciate it if you'd share it with friends, post about it online, or leave a review on Apple Podcasts or Spotify. We welcome your feedback too, either via our website, cognitiverevolution.ai, or by DMing me on your favorite social network. Finally, for now, a couple of quick disclaimers from me and from a16z, which, as you may know, recently acquired the Turpentine Network. First, from me. While I have lots of help in the production process and the business side of the show, for which I'm very grateful, I personally am solely responsible for all content on The Cognitive Revolution, including guest and topic selection, The questions I ask, and the editorial commentary that I offer. For this episode with Matt in particular, I was under no pressure and I received no consideration to have him on the show. And while we did follow our usual practice of sharing questions in advance and allowing our guest to review our edit and request any additional cuts, no topics were off the table and nothing meaningful was cut in the editing process. And second, this is directly from a16z. This information is for general educational purposes only and is not a recommendation to buy, hold, or sell any investment or financial product. Turpentine is an acquisition of A16Z Holdings LLC and is not a bank, investment advisor, or broker-dealer. This podcast may include promotional advertisements. Individuals and companies featured or advertised during this podcast are not endorsing AH Capital or any of its affiliates, including but not limited to A16Z Perennial Management LP. Similarly, Turpentine is not endorsing affiliates, individuals, or any entities featured on this podcast. All investments involve risk, including the possible loss of capital. Past performance is no guarantee of future results, and the opinions presented cannot be viewed as an indicator of future performance. Before making decisions with legal, tax, or accounting effects, you should consult appropriate professionals. Information is from sources deemed reliable on the date of publication, but Turpentine does not guarantee its accuracy. With that, I hope you enjoy this exploration of AI policy from the techno-optimist Little Tech perspective with Matt Perault, head of AI policy at a16z.

Matt Perault, head of AI policy at a16z. Welcome to The Cognitive Revolution.

Thanks so much for having me on.

I'm excited for this conversation. Obviously, AI policy is a hot topic, and I've got a bunch of different angles that I want to get into it with you on. But maybe just for starters, because I think we'll actually have a lot in common here. I noticed in my prep that you are a fellow at the Abundance Institute. And so I wanted to just take a minute to give you the floor to kind of share your dreams for the AI future of abundance and get a sense for kind of what you envision that looking like.

Yeah, so my primary affiliation is as head of AI policy at Andreessen Horowitz, but I have a couple of side things that I had started as an academic before this. I was at UNC Chapel Hill running the Center on Technology Policy. And so as a fellow at Abundance, that's one thing I've been able to continue. I'm also a fellow at the center that I used to direct, which is now at NYU. And those affiliations are awesome because they enable me to kind of continue working with certain communities that I really enjoyed working with in the past. And the Abundance Institute is a great organization. Christopher Kubeman is the lead for it, and it's really focused on a regulatory agenda that can help unlock abundance. And I've learned a lot from that group. They've got a great group of policy professionals, people oriented around economic policy and stuff. They do a lot in energy, which has been interesting and obviously relevant to my current job at Andreessen Horowitz. So I've really enjoyed maintaining a close relationship with that group. The focus for me in the question that you ask, I think, is going to be a little bit disappointing for you. People always ask, tech people, regardless of where they sit in tech, to sort of forecast the future and what does the future of technology look like. I would say that's something that I've never been particularly good at. I'm not an engineer. I'm not a computer scientist. So the people who are really building the tools tend to sit on a different side of the house. I previously worked at then Facebook, now Meta. I was on the policy team, not on the product side. At Andreessen Horowitz, we've got a lot of people who really understand the technology really deeply. That's obviously an important part of my job is doing it as well as possible. But my focus is— on the public policy side. So my orientation to the question is really like, what is the right policy agenda to enable other people to unlock abundance? Like, how do you think about ensuring that the people who can create the tools are able to do that? And that doesn't mean just total deregulation. That doesn't mean being able to do whatever an engineer wants to do. But it does mean trying to avoid regulatory models where the cost of those models, particularly for startups, for little tech companies, Outweigh the benefits. And so that's really the day-to-day focus of my work.

So I hear you on all that. Do you, you gotta have some dreams though, right? I mean, the future of, like, for me, it's, you know, medical advances, equal access for everybody around the world to the sort of AI doctor, the robot maybe one day that's in my home that's like doing my dishes and picking up my kids' messes and stuff like that. Maybe it's one day, like, People don't have to work if they don't want to, or if they don't have work that they find intrinsically motivating or fulfilling.

I don't know.

I'm always interested to hear, especially because you're so invested in this. I would love to hear sort of what animates you.

All those things sound cool. I mean, both my parents and my sister are all mental health professionals. And I think when you talk about AI and medical care, I sort of think about breaking down barriers to access to mental health. Treatment. I mean, that's been something that I think I've had a conversation with about my parents since I was really little. And they were talking to me about the value of having mental health support in your life. And that's hard for people to do for a bunch of different reasons. It can be cost prohibitive. It can be hard to identify a mental health care professional. It can carry a stigma. And so, and so I think some of the advances there are really exciting and interesting. Obviously, it's important that that be done in a way that is consistent with providing a high quality level of care. But I think breaking down access barriers is a really key thing. I've been excited about it as someone who spends a lot of time writing and reading, the benefits that AI can bring to the analytical process and to the writing process. It's a scary one because I think it's coming after my business model. My ability to have some level of writing competence has kind of always been a part of my skill set that has been important in my academic life as a student and then as a professor and been important to my professional life too. And I think the ability of AI tools to develop thoughtful, clear prose is something that corrodes. You have to be a differentiator on top of that. Another way to say it is, at one point when I was working at Facebook, I said to a boss, I understand my job now is to deliver a bad first draft to you. And his response was sort of funny.

He was like, yeah, that's great.

That's great. That's great. And then he was like, but every now and then, could you develop a good first draft? And to the extent that like the jobs that some people in these fields have is to produce that first draft that enables people to provide feedback and critique and then massage and hopefully turn into like a really good final. You know, now we have tools that are able to, I think, do first bad first drafts and maybe at some point do very good first drafts. So I think the fact that it's coming after a thing that has been kind of important to my professional life, I think is like a really interesting, exciting opportunity and also somewhat of a scary one.

Yeah.

But one that I think is exciting because I think it will again play an important role in breaking down access barriers.

Yeah, I feel that very much myself. Coding and even podcasting. NotebookLM, I feel like in many cases competes pretty effectively with what I'm able to do. So yeah, it's coming for all of us.

What is the balance for you on fear versus exciting motivator? As a writer, I think if there's something exciting to feel like, Usually it's a person, now it's a machine and people who are kind of like chomping at your heels. And I think that can result in being more creative and being more focused and devoted to parts of your craft in certain ways. It's also scary to feel like people are potentially coming for your work if you don't perform particularly well. I haven't thought about it as a podcaster. What's the fear versus excitement balance?

I describe myself as classically ambivalent, meaning the dictionary definition of ambivalent is like contrasting or even contradictory, but strong feelings. And that's always been very natural to me with AI. The upside is legitimately thrilling, and I get new models and new products come out. I'm always rushing to go try them and get hands-on experience it today. This morning, it's been Claude for coding up an app and Operator for doing some web tasks. And I do genuinely feel a thrill from these new product experiences. And then I'm less concerned, honestly, about it coming for my work, not because it's not, but because I just kind of worry about even bigger things like what does gain of function AI research look like? And do we really have the wherewithal to control this phenomenon in a macro sense? I'm less worried about meaning or jobs or sort of what are we going to do with our time? I would say I'm like an optimist when it comes to people being able to figure out how to use their time and have a good life, given resources, space, and opportunity to do that. But I'm a little concerned about just like, boy, we don't really know how these things work. We don't really know why they do what they do. And they do seem to be demonstrating more and more of the sort of negatively thrilling behaviours. As much as it's thrilling to see these things go about these tasks and learn to overcome obstacles and whatever. I also can't help but see in that, that paired with deceptive sort of behaviours and scheming sort of behaviours that we are kind of seeing emerge, it's like, geez, I don't know, the dynamics of that just get to be really weird and really hard to predict. But really, the only thing that I kind of don't expect is like a business as usual future. So I try to And obviously I'm a small player in history. I sometimes call myself the Forrest Gump of AI because I always find myself in these notable scenes, but usually as an extra. I just try to make whatever little nudge I can in a positive direction. And I kind of feel like whatever those odds are, it's less for me about handicapping and more about, can I put a little bit of shoulder into shifting the odds of the radically positive future just a bit? And hopefully Hopefully we can all make a little impact like that.

That seems right to me. And I think that feels like more of the direction of travel. I mean, you probably track the macro conversation more closely than I do. I had sort of thought initially in the post-ChatGPT release moment that there was a lot of focus not on the immediate, not on what's exactly in front of you, but on the infinite future, like robots taking over the world. What happens in a world where people have less autonomy? What does that mean philosophically? Like those big picture stuff. Big picture of things. And then over time, it seems like as the technology has improved and improved, more models have become more available to more people. People have a better sense of how to use them. Our digital AI literacy, I think, has improved. People compare notes with their friends about how they use them and they integrate some of those use cases in their own lives. The gaze has turned much closer to people. So kind of exactly as you're describing at the end, More about marginal value creation as opposed to, I'm creating a robot that is going to take over my life. And my sense at least is that when the focus is on the marginal value creation, that has connected people much more closely to a value proposition that's much less scary. And so the question is less like robots taking over the world and more like, I need ideas for recipes, or how do I deal with a kid who's not going to sleep? Or, I'm looking for activities to do on a family vacation in this place, or whatever it is. Or, I'm struggling with how to get this piece off the ground. How can I structure the ideas? Or, I've been using AI sometimes for— I don't like writing conclusions. AI writes very good conclusions, or at least gives you a bad first draft of a conclusion. It can be helpful for that kind of use case. And that to me has shifted the conversation and then also I think shifted Importantly, given the nature of my work, how people think about public policy. Because you want public policy to help us avoid harms, long-term catastrophic harms. But I think you also want policy to give you room to unlock and experience the value. And thinking that through, what are the right models, I think changes as more people have a closer connection to the value proposition.

Maybe that's a perfect segue to jumping into some policy questions. I think folks will know a16z as obviously a prolific investor, super successful investor led by techno optimist thought leaders. And a huge theme of everything that I have read of your work and the firm more broadly is making sure that there's room for little tech as opposed to big tech to experiment, innovate, bring things to market, develop new use cases, et cetera, et cetera. With all that in mind, you said it's not about opposing all regulation, but what is the sort of first order articulation of the policy agenda that you would advocate for today?

Yeah, so I can even frame, I think, what's kind of baked into your question even slightly more explicitly, which is like, a lot of people think that we are extremely focused on being as deregulatory as possible. I would have done my job effectively if we saw no AI policy. That's not the case. And it's not the case, I think, because it's connected in a really important way to the economics of the firm. So the lifecycle of an Andreessen Horowitz fund is 10 years. That's actually significantly longer than the lifecycle of a tech vesting cycle at a public company. tech company. So when I was at Facebook, it's a 4-year vesting cycle. I think it's a longer time horizon than you would typically see in private equity. We're not looking to juice stock prices tomorrow or looking to produce massive returns tomorrow. We're looking to create a healthy, interesting, exciting, vibrant, abundance-oriented, big dream-oriented, to get back to your first question, view of the world and reality in AI technology Over a longer period of time. And I think that's important for a bunch of reasons. One additional way that you could think about it is if people were really into AI tomorrow, really excited about it, and there was an explosion of interest in AI and explosion of AI products, and then the market cratered a year later because of lots of examples of harms in our society, we're not going to get the economic returns that we're aiming to get. We're trying to build healthy, long-run, strong companies over a period of time. And when you're trying to do that, that means that you need to allow startups to get companies off the ground. So you were getting at this in your question. We are not focused on the health of big tech companies. I mean, sometimes big tech interests overlap with ours and are aligned with ours. Sometimes they're not. That's really sort of beside the point. Our focus is for little tech companies. How do they have room to grow? We have to have an ecosystem that's also providing a set of tools that people like to use and feel safe and secure when they're using, when they're using them. And so the kind of core of our policy agenda is focused on trying to encourage policymakers to regulate harmful use, not to regulate AI development. And in our view, that leaves room for little tech companies to build and grow. It's typically how technology regulation has historically been approached. Like if you think about software development, There's not a lot of regulatory intervention in the building of software. But if you create software that's harmful for the world, depending on the specifics of what it is, but in most cases, there's going to be law that's going to be able to take account of that harm and prosecute you and deter you from future activity where you're building something that's harmful. So that's our orientation, is to try to leave the process of development available to little tech, give them room to build, But then importantly, the second component is regulating harmful use. Hey, we'll continue our interview in a moment after a word from our sponsors.

Yeah, so let's dig in on those maybe one at a time. I mean, on the development side, obviously development can mean many things, right? It can mean me sitting down and fine-tuning a model and just tweaking it a little bit for my purposes and baking it into an app. On the very high end, it means pushing the frontier of AI capabilities. And while I'm definitely not a big analogy guy, I do sometimes think of this as being akin to more like biological research. When I think of these sort of, obviously, the canonical example, without coming down firmly on exactly the origin of COVID obviously there are these institutes out there in Wuhan, elsewhere, that are doing this sort of research where they're like, well, geez, what would happen if we made a virus do this? And I think they have good intentions, to be clear. I understand the idea is we're going to make these things so we can figure out how to treat them so that if it ever does happen, we'll be prepared. But then we do have this history of lab leaks. So I sort of think at the very frontier, it does seem like there is a very material risk that is analogous, pretty analogous to the sort of biological research thing where you don't necessarily have to deploy it. You don't even have to intend to deploy it. If you're developing something that's sufficiently powerful, there is kind of this risk that it might get out of control. And so I wonder how you feel about that risk, which I think for most people in the AI safety community is like, Honestly, the number one concern.

Yeah. So I don't know enough about the regulation of biological research to know whether it's a good analogy or not. It feels, at least with my superficial understanding, to be a good one. We give broad latitude, I think, to researchers to research, including things that could be very harmful if they are released in a way that violates the law. And the reason that we do that is there's a lot that you can discover and a lot that can be really beneficial to society from engaging in the research process. The research process itself isn't really good or bad. Typically, again, I think we want to allow researchers a lot of room to explore and not be super prescriptive. There are any number of different things that researchers can do that are problematic. One that's closer to a thing that I've worked on is I was at Facebook when the Cambridge Analytica incident occurred, and that was data going from Facebook to an academic researcher. And so that transfer, at least under Facebook's terms, was legitimate. And then the researcher misused the data, but it was actually The initial transfer to a researcher was consistent with Facebook's terms. And I think in some ways, most ways maybe, you would think that that initial transfer is desirable. We want academic researchers. Typically, the research community is asking for more access to data, not less. That is probably a positive thing. The abuse, misuse of data once you have it in your control is really a problem. And we would want to make sure, I think, that we have stringent rules in place to penalize misuse. And so again, in your COVID analogy, it's not okay to develop a virus that could cause a global pandemic and then release it into the world so as to cause a global pandemic. That I don't believe— again, I don't know the legal regime, but my assumption is that that is unlawful and therefore should be penalized as such. The research, I think, is a different question. And again, mapping it back onto AI development, The development itself is just science. It's math. And so the concern that our firm has had is that when you look to criminalise the math, when you look to make it harder for scientists to do science, that just has the impact of slowing development and actually doesn't really do much to address potential harmful uses. And so if we're concerned about harms, we should try to make sure that they're punished. I think the concern that I have And maybe it's why so many people think we're purely deregulatory. I think people spend a lot of time hearing, thinking, absorbing, don't regulate development. And then when we say, and regulate harmful use, it's like then the ears close up. Because I think maybe the idea is it doesn't feel like there's a clear pathway there. Maybe the view is it's just a diversionary tactic. We're trying to get people away from regulating AI development. So that they do nothing. But actually, we think regulating harmful use is like, there's a lot for lawmakers to do there, lawmakers and enforcers, that is really significant and meaningful. When I was in law school, I spent a summer at the criminal section of the Civil Rights Division of the Justice Department. And that was an interesting experience for a bunch of reasons. Primarily, people go there because they want to litigate cases, I think. And when you're there every day over the course of a summer, what you see is Especially as an intern, you're not traveling around the country going to different courtrooms. You're in the offices in the Civil Rights Division watching what the attorneys are doing when they're not in court. And what they're doing is building cases. And that's really hard to do. Cases don't just come with a clear fact pattern where you can really understand exactly what's going on and where a violation occurs. They don't come to you with that clarity. You have to build the case and you have to identify the violation. I think the same thing will be true in AI. We're not going to know— When AI is used in a violation of anti-discrimination law, that's not going to be like— or I should say it this way, it's not going to necessarily be obvious. You will need to build cases. You need to identify the harm that's occurring and an AI tool being used for that harm. And that, I think, requires work and investment and probably lawmaking in some cases. You need to have enough people to do the enforcement. Those people need to have The resources that they need in order to do the enforcement. They need to have the technical understanding of the technology in order to understand when laws are being violated. There may be cases where existing law isn't sufficient to take account for harms that are created by the use of AI. In those cases, we may need to think about tweaks to existing law to make sure that you can take them into account. So when we say don't regulate development, that means something. And then when we say regulate harmful use, we consider that second component to be a really active part of Let me do one more beat on the development, and then we'll circle back to the harmful use.

Obviously, a big kind of pattern of bills that have been put forward, which I think the proponents would say is their effort to meet folks like you in the middle on this, or even be quite compromising, and also I think that they would say reflects like a genuine shared concern that they don't want to squash frontier research, certainly unintentionally.

Yeah.

HADRN has kind of been recently, and this is now happening again in New York with a bill that I just did an episode on there, the RAISE Act, which I'm sure you're tracking. Basically, the idea is like, we'll have some threshold. I know you have thoughts on thresholds. And if you are above that threshold, and typically they're trying to design the thresholds so that it's like high single digit, maybe low double digit numbers of companies would be caught by these thresholds, then you have to develop a safety plan, publish that safety plan, share it with a regulator, maybe publish it publicly. Those little details vary. Then you have to follow it. And then there's some variants on the provisions of like, do you have to have an audit or do you just self-report or whatever exactly? Right. But My impression is that you don't favour those, but it does seem like that is pretty consistent with the general principle of giving people pretty wide latitude, because most of these proposals, at least the ones that seem to be getting the most traction, are like, the companies get to define their own safety plan and they're basically grading their own homework for the most part. We just want to make sure, speaking on behalf of the sponsors of the bills, that they're actually doing it in a serious way? Because we do think this stuff is pretty serious. And yeah, we're not going to tell every biologist exactly what experiments they can and can't run, but there is a security level 4 type concept where if you're not implementing certain standards, then we look very unkindly on that. So how do you feel about the— we can bring in thresholds here, but also just this kind of very top tier, grade your own homework, but we really want to at least know that you're doing it kind of vibe.

Yeah. So just as a threshold matter, you agree that those regulations are targeting AI development, right? Not harmful use?

Yeah, I think so. Yeah, I would say frontier development is usually the phrase that I—

Yeah. So there are a bunch of ways to get into this. I mean, one, just at the kind of, I guess, top-tier thing is, what is the problem with targeting harms? If what we're concerned about are harms, why don't we put the emphasis on ensuring that we can address harms when they occur. In terms of the development side, which I know is what you want to focus on, I think there are a couple of different aspects of it. One is who do the bills cover? And then the second is for the organizations it covers, what does it ask you to do? And is that— your words were reasonable. I wouldn't use the term reasonable. I think I would say, is that going to encourage the kind of AI ecosystem that we want to see? On the who side, I think you actually, even in your question, you said the thing that it sounds, as you framed it, it sounds like a compromise. It sounds totally reasonable. And it's the thing we fear most, which is this is just going to apply to a handful of companies. 5, 10, 15 companies. It's just going to apply to a couple, a small number. That's the thing that we're worried about. Because the other way to say it is this is going to be a concentrated market and we're going to have a regulatory model that is not only going to Like be okay with a concentrated market. We're actually going to have an approach to regulation that is going to create concentration. Like you can only build at the front. There are only going to be a small number of companies that are able to build at the frontier. That's not what we want. We want to ensure that startups can build at the frontier and that when, if you're a founder and an engineer or like a founder and 5 engineers, founder and 10 engineers, Those kinds of companies are able to have the ambition of building at the frontier. So when you said in your opening, what's the dream of abundance? And I said in sort of a roundabout way, the way that I think about that is what is the regulatory model that allows smart, creative, ambitious, skilled entrepreneurs to build great businesses? This is what I'm talking about. The capacity, the ability to build at the frontier. And it's just going to be 5 to 10 companies line. I thought before I started at Andreessen Horowitz, I kind of thought that was an exaggeration, like was a story that I'd heard people tell, but I didn't think it was a real thing. And now it, I mean, I've heard it a couple of times. I heard it on your previous podcast. I heard it in a conversation with a member of the European Union, someone who's working with them on their AI regulation. And so I don't think it's an exaggeration. I think it's a sort of widely held view that, look, this kind of development is just going to be a small number of companies. And that is, I think, for a lot of people, they'll hear that as very reasonable. And I think it's really important to, whether or not you agree with us, to see it as very strongly opposed to the vision that we would have of what a healthy AI ecosystem would look like. Hey, we'll continue our interview in a moment after a word from our sponsors.

Okay. I, in some ways, am very sympathetic to the idea that concentration of power generally does not appeal to me. So I do find this, for what it's worth, to be the case among many people that would call themselves AI safety-minded people. Broadly speaking, they are like techno-optimist libertarians throughout their lives until this kind of one thing. I think the thing that sort of makes it different for people, that answers the question around what's wrong with just regulating harms, is there may be some things where by the time the harm is done, it's just kind of too late, right? I mean, the COVID pandemic is sort of an instance of that where it's like, and again, I don't take a position on where COVID came from exactly, but if it did come from a lab and now we have 10+ million people dead globally, it's really hard to go to that lab and say, now you have to pay a fine, and act like that's sort of easy enough.

I think it's a perfect example. How many constraints have we put on biological research in the wake of COVID Because as far as I know, the gain-of-function research continues, right?

I mean, I don't think any actual lines of research have really been shut down, have they? But I'm not—

that's what I mean. What if the response to that, instead of focusing on All the problematic incentives and their geopolitical issues and stuff related to how COVID was released into the world, the response was, we're going to make it really hard for every single researcher working on biological research to do their research.

Well, you'd hope to be more targeted and that does obviously connect to the number of companies, but there is something to be said, I think, for, like, I'd love to fix the world. I'd love to have more wastewater monitoring, and I'd love to have air scrubbers installed in all the schools. And there's so many things that seem like, honestly, pretty obvious kind of hardening. And certainly this is a theme in the AI safety discourse too, is like, what can we do to sort of harden society against these potential future assaults? But there is also the notion of choke points. And you can say, okay, there's one thing that seems to be like, a really bad problem. And that is if you create new viruses that don't exist in nature that have super transmissibility or super lethality, that seems really bad, or at least potentially really bad. And it doesn't seem crazy for me for society to come in and say, and I don't think we have, but I would support it if society got together and said, don't create them.

We lived through the pandemic. We didn't create civil or criminal liability for what you're describing. I'm not a biological researcher, so I have no I don't have a sense of why we didn't do that and why we as a society might think that risk is one worth bearing. But my guess is because that kind of research on some of the most virulent, potentially problematic viruses is the same thing that will prevent those viruses from spreading over time. You have to do research on them, I assume, in order to figure out how to combat them. That doesn't mean it's okay to release them into the world. And it doesn't mean that they're— I think there are lots of reasonable things that you could do in that situation to try to ensure that there's a deterrent for the harmful use. But we don't take that kind of research off the table because of the upside. And I think that when we were talking about AI use cases, there are an enormous number of positive use cases from the mundane to the revolutionary. I mean, mundane might be, it helps me write a conclusion. But dramatically lowering barriers to medical access or mental health treatment or making it much easier for more developers to build more products that are valuable to people, whatever the explosively positive use case is, I think that just like with biological research, the reason we preserve a fair amount of ability to do the research itself is because of those possibilities of really positive explosive use cases. And I think if what we want is abundance, and I think there are a lot of people who are skeptical of abundance, but if that is the thing that we want, then we need to have a regulatory environment that is conducive to it. And putting the significant constraints, handcuffs on the development process, I think is not going to get us there. And there are a few different components of it that feel to me to be particularly off base. One is just going to apply to a couple of companies. That is another way. I mean, For us, that means we can't invest in startups to grow. I also think it means that's one part of the innovation angle. The other part of the innovation angle is one of the reasons that our laws disfavor monopoly is that monopolists tend to not innovate rapidly. And so if you have a super concentrated market, it's likely that you're going to see fewer of those most interesting use cases be pushed and be pushed aggressively. The other thing which I think is really important is it would be— I think it would be a different story if we said, okay, we're going to make it a lot harder to build AI models, but the impact of that is going to be that we avoid a significant number of safety incidents and we really can ensure that we are going to dramatically reduce harm in the world. And I don't think that's the case. I don't think that publishing a document with safety protocols, that doesn't mean that you're going to not have safety incidents. I think there are probably some number of safety incidents that wouldn't occur that might otherwise occur, but I think it's conceivable that there are going to be very few of those, that it actually has no direct correlation to actual safety of the performance of your product. And if that's the case, if there's pretty limited efficacy in what you're doing, then you're simply burdening development without a lot of upside. And that again is why we think The idea isn't don't regulate. The idea is put resources and emphasis on the harmful use side, both because it leaves room for development, but also because it actually, I think, would be more effective in addressing potential misuse.

I do agree with you that the proposals that have been put forward of this sort of like everybody has to have— not everybody, of course, some number, which we could debate exactly how big that number might grow over time, but that these companies above a certain threshold have to develop a safety plan and follow it and disclose safety incidents. I would agree that's not that strong of a barrier against certain, whatever possible bad things happening. I think it is more meant to try to make sure that we are not flying blind. And that is honestly one of my bigger worries. I mean, I wonder if a reframing of some of these requirements around transparency is appealing at all. One of my worries is that, and I think we are seeing signs of this from OpenAI and others, and frankly, Anthropic, who has, at least in some corners of the safety community, been the sort of darling frontier developer. They seem to be going for some sort of self-improving feedback loop, some sort of recursive dynamic, Where it's like, AI is going to get so good at coding and ML research that it's going to do the improvement. And then literally somebody from Anthropic, as they launched Claude 4, said, we want Claude to take over all the ML research so we can all go to the beach. And people were like, well, you can't put it much plainer than that, right? I do wonder what starts to happen behind those closed doors. And again, it's all very speculative right now because we're just in unprecedented territory all over. But I also think Daniel Coccatello, who is one of the lead authors on the AI 2027 projection or forecast or scenario, I guess might be the best word for it. One of the things that he says is that we should expect— right now we have a very narrow gap between frontier AI capabilities and what is deployed for you and me to use. But he expects that gap to grow much wider over the next couple of years as the companies realise that And I think they maybe already know it, but that they're in this sort of race for supremacy. Who can create the most powerful AI model first that dominates the market, dominates their rivals, what have you, and that they just aren't going to want to show that to anyone. And right now there's not really any rule that would require them to do that. So I wonder if you would support something that's like transparency requirements, just even if only around what capabilities the AIs have. OpenAI, Anthropic, Google, you have to tell us what is it that your AIs can do so we at least know what they are capable of in your labs. What do you think about something like that?

Yeah, so we published a blog that I'd encourage anyone who's interested to read because it's a little hard to sort of run through it quickly. But the basic idea is that a lot of the transparency proposals we think are probably unlawful. They probably violate the First Amendment and also aren't particularly useful for consumers. So I don't think some of the things that have been proposed, some of the transparency mandates around Like certain kinds of assessments of safety risk are actually things that will provide consumers with information that's useful for them, that's really informative and shapes how they interact with the model or which models they choose to interact with. And in our view, the way to get at a disclosure regime that's lawful, that is useful for consumers, and that is also not unduly burdensome for little tech is something that we call AI model facts. And so it would basically be a basic fact sheet that would run through different information that would be helpful for users. And this is sort of a fun project to work on because it was interesting to see, like, in current disclosure mandate proposals, what are the things that might sort of map onto this? Like, what would actually be, if the criteria is not unduly burdensome for startups, useful, not unlawful, legal, What are other elements from current proposals? And there weren't that many actually that I'd say would meet that criteria. The kind of stuff that we thought of were things like knowledge cutoff date, which I think is actually extremely useful for consumers. So if you ask an LLM to give you information on who won the basketball game last night, that's not in its training set. So having a sense of exactly when that cutoff date is will enable you to ask A, to ask better questions that are actually within the corpus of data, but B, if you were to ask it who won the basketball game last night and it gave you an answer, the likelihood that that's a hallucination or information that you shouldn't trust should be high. So I think that's the kind of information that actually is useful for consumers. The courts have been pretty clear that it's very hard to survive legal scrutiny if you require information that's either controversial or not factual or burdensome for companies. And so I think some of the types of information that are typically seen in transparency proposals, like speculating about potential long-run harm or forecasting potential safety risks, I think it's going to be really challenging for those things to survive First Amendment scrutiny.

One more quick bit on the thresholds. Again, share your— without going on too long, share your concern about regulatory capture and the sort of structural oligarchy of various markets that we could find ourselves in via mechanisms like that. I think the people who have developed these proposals around thresholds have come to them because they see the scaling law phenomenon as a good indicator that we're kind of headed there.

Regardless.

And certainly it seems like we basically are there today, right? I do a live players analysis every so often where I'll ask people, who do you think should count as the top tier AI companies in the world today? And the list is sometimes as short as 3 that I get back and never longer than 10. And that seems to be, at least so far, a structural reality of just the kinds of inputs that these things require. And yes, the costs are definitely coming down. And so people have sort of proposed naturally escalating thresholds, whether it be like a certain amount of money or even could be revenue, which I know you're a little bit more favourable toward than other concepts. But I guess, do you see the notion that there's going to be a lot of companies at the frontier of AI seems to run contrary to what increasingly seems like it is kind of a natural law, that there is just sort of a scaling law reality. And as you know, as a venture investor, right, power laws rule everything around us. So is it so unrealistic to think that it could stay a small number of companies caught by these thresholds?

Well, so first of all, I think if what you're trying to do is really just capture large companies who are able to bear compliance costs, revenue thresholds are clearly the policy vehicle that gets you there. And the reason is, once you're taking hundreds of millions of dollars in the door, we can figure out what the threshold should be. I've seen proposals like $100 billion thresholds. I've seen higher revenue thresholds. I've seen lower ones. But whatever it is, at some point of taking revenue in the door, you are able to devote some percentage of that to compliance costs.

Right.

It's hard to argue if you If your annual revenue is $500 billion, that you can't devote some percentage of that to whatever kinds of exercises and transparency disclosures regulators want you to do. Compute and training cost thresholds are different because we think startups can build at the frontier. They are able to build models with significant compute power. And I had a conversation with someone on our technical team about a $100 million training cost threshold. Because I was starting to see that in more pieces of legislation. And I said, oh, is that a good way to carve out little tech companies? And he said, and I trust him on this, I don't have the technical chops to critique it, but he said every model costs $100 million to build. And so I think those thresholds are not going to be successful in separating out little tech from big tech. And the technology will outpace the ability of regulators to update them over time. One thing I noticed in the RAISE Act, and as part of the discussion that you had about it, is that there are a couple of different ways you can hit the large developer threshold. You have to hit a $100 million training cost cap, but I don't think there's anything that says that that couldn't be cumulative, which means that if you built a large number of $25 billion models, you'd be captured. Right now that might seem very far in the future for a startup, but you might have startup that's building those models a couple a year. I don't know exactly what the highest rate of model development might be, but more than one a year over a period of time. And at some point, that would mean every startup is a large developer. And so those kinds of thresholds are focused on the development layer, which is where we think regulation is really much more fraught. If it's cost or compute, that's focused on the science of building the tool. And really doesn't have anything to do with the ability of a company to handle the kind of regulatory complexity that some of these regulatory models require. Another way to get at your question, I think, too, is I think you're right that there are these natural barriers to entry in AI. So cost of compute. And we have some ideas actually about different ways to try to provide broader compute Resources to more entities. Access to data, we have some ideas there about trying to ensure that more startups have more data that they can train on. The question for policymakers is, do you want to add an additional significant hurdle? Do you want regulatory barriers to make it even harder? And if that's the case, then we end up in the world that I think you're describing, which is we have a small number of developers who can build at the frontier. Everyone else can have the rest of the ecosystem, less capable models, not building at the frontier, but we essentially have a regulated monopoly that looks more like what we've had in telecommunications, for instance, a very small number of companies with stringent level of regulatory oversight. At one point, even like an explicit government grant of a monopoly to one telecommunications provider. And I think there are people who would say that's the right model. That's exactly the direction of travel that we should I think from our standpoint, that is a level of monopolization that we would think of as problematic. It, I think, suggests a lower level of innovation than we would think is desirable and reduces the ability of startups to compete at the frontier in a way that we think is going to unlock a lot of value for people.

One thought I had earlier on the 10-year time horizon of an a16z fund is that not only is that longer than vesting schedules, it's also longer than most people's AGI timelines these days. And so I wonder, something I've pitched many times and it never seems to get any traction, but there is sort of a version of it right now that's under consideration in California. If you squint at it with SB 813, is some ability to sort of move in and out of different regulatory regimes. Like, the simplest possible thing might be like, have a sunset clause on a lot of these things. Like, okay, for the next 3 to 5 years or whatever. We think we're in a critical period. Let's put this in place, but then have it sunset at 3 years. And if it turns out some of this stuff was hype or we got certain thresholds wrong, then they can just sort of disappear. With SB 813, there's kind of a similar concept where it's like a trade is able to be made between an AI company that wants to have a develop— a liability shield if they opt into a certain regulatory scheme that's approved by the state but administered by a private organisation, then they can have that liability shield, but then they could also potentially opt out of it at a future point in time if it's not a trade that's working for them for whatever reason anymore. So I guess 2 questions there. One is, what do you think about SB 813 specifically? And do you have any other creative ideas for ways that we can put some measures in place, whether they be guardrails or transparency or whatever, that don't lock us in long term, because nobody wants the GDPR of AI. But definitely some of us are nervous about not—

A lot of lawmakers do seem to want the GDPR of AI, because I think the proposals that they are moving forward with suggest that. I mean, one sort of interesting thing is I saw a story, and I don't know how accurate it is, but I saw a story that the EU is considering pausing implementation of the EU AI Act. So the EU is even having concerns about its own AI-oriented version of the GDPR and starting to pull back. The only state that's actually enacted legislation to govern AI comprehensively is Colorado. And the governor of Colorado came out in support of the federal AI moratorium, which would have made it impossible for him to essentially enforce his own law. So policymakers who have actually enacted these approaches are, I think, Expressing some rightful concern about how they'll be implemented in practice. Again, that doesn't, from our standpoint, mean that policymakers shouldn't take action. We think lawmakers should focus on actively regulating harmful use. We think there's a role for states and the federal government to play in that. Regulating harmful use is actually pretty consistent with how states historically, and according to the Constitution, have gone about lawmaking. So they can't unduly burden interstate commerce, but they can police harmful conduct within their jurisdictions. And some areas of law are disproportionately occupied by state governments. So like most of criminal law is at the state level. So if you're looking to tighten criminal law to address harmful criminal use of AI, most of that is really for state lawmakers to do. Investments in workforce development and AI literacy. There's a lot for states and the federal government to do there.

Transparency.

We've talked a little bit about a model facts, AI model facts model. So I think there's a ton of stuff. that lawmakers and enforcers can do to try to arm us for a world where AI is more prevalent. That doesn't mean that there's an infinite road. The way that you described it, I think of like, hey, could we just do some things for a short period of time? And if it's not the right approach, we can pivot. I like that kind of experimentation generally. But I do think depending on the model, like if it's a really stringent regulatory model, it's kind of like saying, We want to run this race, we want to run it as quickly as possible, but you're going to run— you're going to have to wear a 20-pound backpack for the first mile. And I don't think— again, different people have different views of like how important it is to run that race at a certain pace, but there— it's not cost-free even to do it for the first mile to require someone to wear a heavy weight.

Yeah. So where does that leave you on SB 813 and the sort of quasi-private, at least more dynamic regulatory idea there?

Yeah, I mean, it's something that we're still exploring. I think there are lots of promising things about it. The idea that lawmakers will have to wrestle with tort liability at some point, I think, is important. I think that's kind of baked into the fundamental premise of the bill. And I haven't seen that many proposals that really seek to do that. And I think if we want to see people have even very mundane uses of AI in their lives that will improve their lives in some marginal but also level that is meaningful, you have to think about tort law, I think. And so I think the fact that Bill does it is a positive thing. I think the question is whether there's a regime there that works for little tech. And I think there are some things that people have said about it that don't fully wrestle with that component. So some people say, well, this is just voluntary, but immunity from tort law or some level of immunity, some protection from tort law is a massive benefit. And so I think if you said, look, you can get immunity from tort law, but you have to pay $20 billion, most people would say that doesn't sound fair. That doesn't sound workable for little tech. It means every large company is just going to pay to get immunity from Tort law, and every small company is going to have to bear that legal risk. That's not really voluntary. You're saying there's this very valuable benefit and you're making it prohibitive for startups. So I think the question is, is the regulatory regime in 813 one that's workable for startups? And I think there are elements of it that are, but there are elements of it that aren't. My hope is that over time, as that bill is examined, that it comes into a form that's more workable for little tech. And if it's workable for little tech, that's the direction of travel that we want to see. that gets us closer to support.

Do you want to leave the audience with any other thoughts, which could be any maybe specific things that you're tracking and supporting right now, or just other priorities or ideas that we haven't touched on and want to make sure people are aware of?

The one thing I was going to flag is the Trump administration is releasing this National AI Action Plan at some point in June, July. So that's like a thing to look out for.

Yeah. Well, I definitely expect that there will be many more developments in this story before we get any sort of stable resolution or stable policy regime, let alone, I mean, even stable technology regime. So I appreciate this. I think a good constructive conversation, and I hope we can do it again because, if anything, all the time intervals are getting compressed because I suspect all these things are kind of tightening and going to be coming at us with a level of intensity that, if I'm generally right about the direction of where the technology is headed, is going to require all of us to be coming together and really trying to synthesise all the different perspectives into the best possible plan. So come back and let's do that again before too long. How's that sound?

Sounds great. Thanks a lot.

Cool.

Love it. Matt Perault, head of AI policy at a16z. Thank you for being part of the Cognitive Revolution.

Awesome. Great. Thanks for having me on.

It is both energizing and enlightening to hear why people listen and learn what they value about the show. So please don't hesitate to reach out via email at tcr@turpentine.co.

Or you can DM me on the social media platform of your choice.